Addressing topology specific replicated bus units
A technique for handling cache-inhibited operations in a data processing system includes receiving, at a topology specific replicated bus unit, a cache-inhibited (CI) operation that is scope limited. The replicated bus unit determines whether an address associated with the CI operation matches an address for the replicated bus unit. In response to the address associated with the CI operation matching the address for the replicated bus unit, the replicated bus unit processes the CI operation based on the scope being limited to that of the replicated bus unit. In response to the address associated with the CI operation not matching the address for the replicated bus unit, the replicated bus unit ignores the CI operation.
1. A method of handling cache-inhibited operations in a data processing system, comprising:
receiving, at a topology specific replicated bus unit, a cache-inhibited (CI) operation that is scope limited;
determining, by the replicated bus unit, whether an address associated with the CI operation matches an address for the replicated bus unit;
in response to the address associated with the CI operation matching the address for the replicated bus unit, processing, by the replicated bus unit, the CI operation based on the scope being limited to that of the replicated bus unit; and
in response to the address associated with the CI operation not matching the address for the replicated bus unit, ignoring, by the replicated bus unit, the CI operation, wherein the scope includes a first scope limited to a processing cluster, and wherein the data processing system includes multiple processing clusters that each include a processing cluster level topology specific replicated bus unit assigned to a same first address.
2. The method of claim 1 , wherein the scope includes a second scope limited to a processing node, and wherein the data processing system includes multiple processing nodes that each include a processing node level topology specific replicated bus unit assigned to a same second address that is different from the first address.
3. The method of claim 2 , wherein the first scope is specified by a first value in a scope field of a first base address register (BAR) that stores the first address and is located in a storage subsystem and the second scope is specified by a second value in a scope field of a second BAR that stores the second address and is located in the storage subsystem.
4. The method of claim 1 , wherein the CI operation is a CI load operation or a CI store operation.
5. The method of claim 1 , wherein the replicated bus unit is an interrupt controller.
6. The method of claim 1 , further comprising:
in response to the address of the CI operation matching a first address stored in a first base address register (BAR) of a storage subsystem, issuing, by the storage subsystem, the scope limited CI operation on a first bus that couples processing units of a processing cluster to each other and to the replicated bus unit.
7. The method of claim 6 , further comprising:
in response to the address of the CI operation matching a second address stored in a second base address register (BAR) of the storage subsystem, issuing, by the storage subsystem, the scope limited CI operation on a second bus that couples processing clusters of a processing node to each other and to another topology specific replicated bus unit.